DESCRIPTION (From the Applicant's Abstract): The neural cell adhesion molecule L1 is found on some classes of migrating neuronal precursors in the developing nervous system and on almost all projection axons in both the central nervous system and peripheral nervous system. Not surprisingly, it has been implicated in the fasciculation of axon bundles and in migration of some neural precursors in various in vitro systems. In the early 1990's it was shown that mutations in the L1 gene in humans cause severe mental retardation (corpus callosum hypoplasia, adducted thumbs, spastic paraplegia, and hydrocephalus). We have analyzed individuals with different mutations in the L1 gene and discovered that mutations that lead to a loss of L1 expression are much more severe than mutations that only alter the cytoplasmic domain of L1. However, mutations of the cytoplasmic domain are sufficient to cause axon guidance failures and mental retardation. Recently, we and others have analyzed the L1 knock-out mouse and discovered that it has a phenotype remarkably similar to humans with X-linked hydrocephalus. This includes hydrocephalus, abnormal development of the corticospinal tract, and hypoplasia of the cerebellar vermis and corpus callosum. In this project we propose to test the hypothesis that L1 mediated adhesion is essential for normal development of the cerebellar vermis and that the function of the L1 cytoplasmic domain is essential for development of the corticospinal tract. To do this we will generate new mouse lines with specific alterations in the L1 cytoplasmic domain. We will also analyze mice in which the 6th Ig domain of L1 has been removed, deleting the RGD sequence in L1, allowing us to evaluate the difference between L1 homophilic binding and L1-integrin interactions during brain development. Finally, we will undertake the first careful analysis of cerebellar development in mice with altered or absent L1.